Psychobiology of Suicidal Behavior in Borderline Personality Disorder (BPD) is the first prospective study of clinical, psychosocial and biological risk factors for suicidal behavior in patients with BPD. The overarching goal of this work is to identify risk factors and clinical subtypes predictive of medically serious suicidal behavior. BPD is a highly prevalent psychiatric disorder, found in up to 2% of the population, and is the only psychiatric disorder defined, in part, by recurrent suicide attempts. With a suicide completion rate of 3%-10%, BPD is a clinically relevant model for the study of suicide. This study follows the stress-diathesis model of suicidal behavior, which predicts an increase in risk of suicide when acute stressors, such as interpersonal crises and depression, interact with personality traits such as impulsivity or affective instability. These personality traits are mediated by speciic brain networks and contribute a biologic diathesis to suicide under stress. At intake, subjects participate in a multidimensional assessment of risk factors associated with suicidal behavior. Systematic follow-ups are then conducted at 3 months, 6 months, and annually to define prospective predictors of suicidal behavior in short term (12 month), intermediate (2-4 years) and long term (6+-10+ years) follow-up intervals. Currently, 245 BPD subjects are enrolled, with a 92% trace rate. By 2 year follow-up, 25% of subjects re-attempted suicide. Using trajectory analysis, a High Lethality BPD subtype has been defined by the increasing lethality of their recurrent attempts. A recently completed 6 year follow-up analysis of 90 subjects identified poor psychosocial function as the most consistent predictor of suicidal behavior across all time intervals. This proposal will add 100 additional BPD subjects and 5 years of follow-up to increase statistical power of our analyses in all intervals, especially the long-term interval, whe many subjects will enter the age of highest risk for suicide. Additional assessments have been added for characterizing quality of life, stability of diagnosis and core dimensional traits, as these variables impact both psychosocial and suicidal outcomes. This study also investigates the neurobiologic diathesis to suicidal behavior using fMRI protocols, specifically targeting the neural basis of regulatory functions critical in responding to stress (e.g., response inhibition, conflict resolution, & autobiographic memory). A specific neural network is proposed as a mediator of suicidal behavior. fMRI protocols will test the functional connectivity of key structures in this neural network during negative affective stimulation. Preliminary studies indicate that negative affective stimulation produces interference with cognitive processing in these important regulatory functions. Affective interference increases the risk of suicidal behavior. The results of the longitudinal studies are immediately applicable to guide suicide risk assessment and treatment strategies; while the fMRI substudy may provide a biomarker identifying patients at highest risk for suicide, and a non-invasive method for following the progress of treatment.